JPH10150209A - Photoelectric conversion element - Google Patents

Abstract

PROBLEM TO BE SOLVED: To raise the characteristics, such as an open-circuit voltage, a short-circuit photocurrent, a low-illuminance open-circuit voltage and a leakage current, and durability of a photoelectric conversion element by a method wherein the photoelectric conversion element is constituted into such a structure that the surface of a lower conductive layer has a roughened shape, an i-type layer contains columnar crystal grains and the longitudinal directions of the columnar crystal grains are arranged slantingly to the normal direction of a substrate. SOLUTION: A photoelectric conversion element has a substrate 101, a lower conductive layer 102, a first doped layer 103, an i-type layer 104, a second doped layer 105 and an upper conductive layer 106. Such the photoelectric conversion element is constituted into such a structure that the surface of the layer 102 has a roughened shape, the layer 104 contains columnar crystal grains and the longitudinal directions of the columnar crystal grains are arranged slantingly to the normal direction of the substrate 101. For example, the surface roughness Ra of a length of the extent of several tens of microns of the surface of the layer 102 is set in a length of more than 0.1μm to less than 1μm. Moreover, the microscopic regions, where an angle G120 that the normal of the surface of the layer 102 forms with the normal of the major surface of the substrate is more than 15 degrees to less than 45 degrees, on the surface of the layer 102 is set so as to occupy 80% or more of the entire surface region of the layer 102.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a non-single-crystal photoelectric conversion device in which the surface shape of a lower conductive layer, the crystal structure of an i-layer, and the structure of a doped layer are improved.

[0002]

2. Description of the Related Art Hitherto, studies have been made on improving the photoelectric conversion efficiency and light degradation of a photoelectric conversion element using a pin junction of a non-single-crystal semiconductor.

It is known that, by increasing the dopant concentration in a doped layer, the activation energy of the doped layer decreases, the built-in potential of the pin junction increases, and the open-circuit voltage of the device increases.

It is known that photodegradation is improved by using a microcrystalline material for the i-type semiconductor layer.

In solar cells using microcrystalline silicon (μc-Si), IEEE WCPEC; 1994 Hawaii p409 "INTRINSIC MICR
OCRYSTALLINE (μc-Si: H)-A PROMISSING NEW THIN FIL
M SOLAR CELL MATERIAL ", J. Meier, A. Shah), a photoelectric conversion efficiency of 4.6% was obtained by a plasma CVD method using VHF (70 MHz). Furthermore, a stacked solar cell of amorphous silicon and microcrystalline silicon was fabricated, and an initial photoelectric conversion efficiency of 9.1% was obtained.

It is also known to provide a transparent conductive layer between a substrate or a metal layer and a semiconductor layer. This prevents the elements of the metal layer from diffusing or migrating into the semiconductor layer, thereby preventing the photoelectric conversion element from shunting. Further, by having an appropriate resistance, a short circuit due to a defect such as a pinhole in the semiconductor layer is prevented. Further, irregularities on the surface increase irregular reflection of incident light and reflected light, and extend the optical path length in the semiconductor layer.

[0007]

However, a solar cell using the above-mentioned microcrystalline silicon-based material has a photoelectric conversion efficiency of 4.6%, which is still not practical. Also, a-Si /
The μc-Si stacked solar cell has an initial photoelectric conversion efficiency of 9.1%, but has a problem in that the a-Si layer on the light incident side is greatly deteriorated by light. Furthermore, the thickness of the μc-Si layer was 3.6
Since the thickness is as thick as μm and the deposition rate is as low as 1.2 A / sec, a layer formation time of about 8 hours is required, which is not industrially practical.

[0008]

In a photoelectric conversion device having a substrate, a lower conductive layer, a first doped layer, an i-layer, a second doped layer, and an upper conductive layer, the surface of the lower conductive layer has an uneven shape. Wherein the i-layer contains columnar crystal grains, and the longitudinal direction of the columnar crystal grains is arranged to be inclined with respect to the normal direction of the substrate. Numerically defined, the straight line A passing through the columnar crystal grains and parallel to the longitudinal direction thereof, the shortest distance between the interface 1 between the first doped layer and the i-layer, and the interface 2 between the second doped layer and the i-layer. The ratio of the straight line connecting the distance to the straight line B passing through the columnar crystal grains A being 20 degrees or less is set to 70% or more with respect to the entire volume of the i-layer.

Further, a third doped layer, a second i-layer, and a fourth doped layer are provided between the second doped layer and the upper conductive layer, and the second i-layer is non-conductive. 2. The photoelectric conversion element according to claim 1, comprising a crystalline silicon-based semiconductor, and having a film thickness of 0.1 μm or more and 0.4 μm or less.

The first and / or third doped layer has a laminated structure of a layer made of a microcrystalline silicon-based semiconductor material and a layer made of an amorphous silicon-based semiconductor material; The layer made of is in contact with the i-layer and is a photoelectric conversion element.

According to the photoelectric conversion device of the present invention, characteristics such as photoelectric conversion efficiency, open-circuit voltage, short-circuit photocurrent, low-illuminance open-circuit voltage, and leak current of the photoelectric conversion device can be improved. It can also improve the outdoor exposure test, mechanical strength, and durability in long-time light irradiation. Further, the cost of the photoelectric conversion element can be significantly reduced.

[0012]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1a is a schematic sectional view of a photoelectric conversion element of the present invention, wherein 101 is a substrate, 102 is a lower conductive layer composed of two layers, a reflective layer 102a and a transparent conductive layer 102b, and 103 is A first doped layer made of a non-single-crystal silicon-based semiconductor material, 104 is made of a microcrystalline silicon-based semiconductor material, a layer having i-type conductivity, and 105 is made of a non-single-crystal silicon-based semiconductor material. This is a layer having conductivity opposite to that of the doped layer. 1
It has a function of generating a photoelectromotive force by forming a pin junction with the layer configuration of 03-104-105. 103, 104 and 105 are collectively referred to as a photovoltaic layer. 106 is an upper transparent electrode, 107 is a current collecting electrode.

The lower conductive layer 102, the i-layer 104, and the
The one doped layer 103 has the following features.

(1) The direction and mechanical strength of the columnar crystal grains,
The relationship between photoelectric conversion efficiencies was investigated. As shown in FIG. 12-b, a straight line A passing through a certain columnar crystal grain A and parallel to its longitudinal direction,
The interface 1 between the first doped layer and the first i-layer, the second doped layer and the first
The frequency distribution of the angle F formed between the straight line connecting the shortest distance to the interface 2 of the i-layer and the straight line B passing through the columnar crystal grains A was examined.
And the angle F is 20 degrees or less.
The relationship between the ratio K to the total volume of the i-layer and the photoelectric conversion efficiency after the “torsion test” was examined, and the results were as shown in FIG. That is, it was found that when the ratio K was 70% or more, good photoelectric conversion efficiency was exhibited, and when the ratio K was 70% or less, the open-circuit voltage was lowered, and the photoelectric conversion efficiency was lowered.
The “torsion test” is based on JIS C8917 for crystalline solar cells.
Compliant with paragraph A-10. The conditions were such that a twist of height h = 5 mm was repeated 50 times on an area of 10 cm × 10 cm.

(2) The lower conductive layer 102 of the present invention has a surface roughness Ra of 0.1 μm at a surface length of about several tens of microns.
Above, it is 1 μm or less. Then, (surface roughness) x (first i
The refractive index of the layer and the wavelength of the visible light or the infrared light become substantially the same, and the light confinement effect is exhibited, and the short-circuit photocurrent of the photoelectric conversion element is dramatically improved.

The long-wave light that has not been completely absorbed inside the first i-layer is reflected by the lower conductive layer and reenters the i-layer. However, the light is scattered on the surface of the lower conductive layer. No interference occurs, there is no region that strongly absorbs light, and light degradation can be further suppressed. Similarly, since there is no region that hardly absorbs light, the open-circuit voltage can be improved.

The relationship between the surface roughness and the photoelectric conversion efficiency was examined.
FIG. 6 shows the results. It was found that excellent conversion efficiency was exhibited in the above range of surface roughness.

(3) An area where the angle G (120) between the normal to the surface of the lower conductive layer and the normal to the main surface of the substrate in a micro area of about several hundred angstroms is 15 degrees or more and 45 degrees or less. By setting it to 80% or more of the entire surface area, the thickness distribution of the first i-layer is small, and since there is almost no extremely thin area, the leak current is small, and thus the open-circuit voltage can be increased. . Further, the light confinement effect is further exhibited.

The relationship between the distribution of the angle (angle G) between the normal to the minute region of the lower conductive layer and the normal to the main surface of the substrate and the photoelectric conversion characteristics of the solar cell after the “torsion test” was examined. angle
G is shown in FIG. FIG. 7 shows the relationship between the angle G and the photoelectric conversion efficiency after the “torsion test”. 80% of angle G
Below, it was found that the shunt resistance was reduced (weak short-circuit state) and the photoelectric conversion efficiency was reduced.

(4) The photocarriers generated in the i-layer are moved by an internal electric field, and the internal electric field is substantially parallel to a straight line connecting the shortest distance between the first doped layer and the second doped layer. Therefore, in the present invention, the longitudinal direction of the columnar crystal grains contained in the i-layer is substantially parallel to the straight line connecting the shortest distance between the first doped layer and the second doped layer, and the longitudinal direction of the columnar crystal grains is When the length is not less than 100 Å and not more than 0.3 μm, the chance of passing through the interface between the crystal grains is reduced, so that the fill factor and the short-circuit current are improved.

Further, since the chance of carriers passing through the interface is reduced, the recombination speed of carriers can be suppressed. Therefore, light deterioration can be further suppressed. Further, since the directions of the columnar crystal grains are substantially aligned, the interface state of the columnar crystal grains is small. Therefore, the open-circuit voltage can be improved.

Further, it has a higher light absorption coefficient than single crystal silicon, and has a longer wave light absorption coefficient than an amorphous silicon semiconductor material. Therefore, since long-wave light (infrared light) is effectively absorbed, a sufficient short-circuit current can be obtained even with a film thickness of about 3 μm.

Since the region between the columnar crystal grains is occupied by a good amorphous silicon-based semiconductor material containing hydrogen, there is almost no probability that optical carriers are trapped in this region.

In general, the longitudinal direction of the columnar crystal grains generally has an angle of 10 ° or more and 50 ° or less with respect to the normal to the main surface of the substrate. It is. Therefore, when the roll-to-roll method is performed, even if the long substrate is wound on a roll, the film does not peel off. Since film peeling does not occur, it can be formed on a substrate having a curved surface. Similarly, it is easy to use the photoelectric conversion element of the present invention formed on a planar substrate by bending it. In particular, when the photoelectric conversion element of the present invention is used as a solar cell, it can be used on a curved flat surface such as a wall surface of a building.

Further, since the short-circuit current can be improved as described above, the first i-layer can be made thinner, so that light deterioration, productivity can be improved, and power cost can be reduced.

(5) The i-layer is characterized in that minute regions composed of an amorphous silicon-based semiconductor material are present in a volume ratio of 50% or less with respect to the entire region of the i-layer. therefore,
The open-circuit voltage can be improved as compared with a photoelectric conversion element in which the entire region of the first i-layer is made of a microcrystalline silicon-based semiconductor material.

Although the reason is unknown, the open circuit voltage can be further increased because the leak current can be reduced. Further, the entire area of the first i-layer has improved resistance to external force as compared with a photoelectric conversion element made of a microcrystalline silicon-based semiconductor material. Since the flexibility of the amorphous Si-Si network is superior to that of the microcrystalline Si-Si, the region composed of the amorphous silicon-based semiconductor material contained in the first i-layer has an external force. It is thought to be effective for alleviation of Further, it is considered that the internal stress is similarly reduced.

When the volume ratio is 50% or more, light deterioration becomes large. Therefore, it is desirable to use a stacked element structure having a structure such as a-Si / μc-Si.

(6) In the case of a stacked cell as shown in FIG. 1B, the third doped layer 11
0, the second i-layer 111, and the fourth doped layer 112 are sequentially laminated, and the thickness of the second i-layer 111 is 0.1 μm or more and 0.4 μm or less. As described above, in order from the light incident side, the second semiconductor material having a large absorption coefficient of short-wave light such as a-Si is used.
By forming the first i-layer with a semiconductor material having a large absorption coefficient of long-wave light such as μc-Si, the spectral sensitivity to light in a wider wavelength range can be improved. Can be increased.

Further, the open-circuit voltage can be higher than that of the photoelectric conversion element in which the first i-layer is made of μc-Si, and the photoelectric conversion efficiency can be improved. Also, the spectral sensitivity is different
By connecting the i layers, the second i layer made of μc-Si can be made thinner, so that the fill factor (fill factor, FF) of the photoconductive property can be improved.

Further, since the second i-layer of the present invention has a thin film thickness of 0.1 μm or more and 0.4 μm or less, even if an amorphous semiconductor is used for the second i-layer, light deterioration can be suppressed as much as possible. is there.

Further, although the reason is not clear, by laminating the second i-layer made of an amorphous silicon-based semiconductor material, the leak current of the photoelectric conversion element can be reduced. Therefore, the open-circuit voltage can be further increased.

[0033] In particular, the photoelectric conversion element of the present invention is an optical sensor,
When used as an image sensor, it is important to reduce leakage current. In addition, when the photoelectric conversion element of the present invention is used as a solar cell, a high open-circuit voltage can be output even under low-illumination irradiation light, so that, for example, power generation efficiency can be improved even on a cloudy day, morning, or evening. It does not fall extremely.

As described above, since the lower conductive layer is not planar, light does not interfere inside the second i-layer, so that there is no region that strongly absorbs light and light degradation is further suppressed. be able to. Similarly, since there is no region that hardly absorbs light, the open-circuit voltage can be improved.

(7) Method for Forming First i-Layer The i-layer 104 is formed by a plasma CVD method using an electromagnetic wave having a frequency of 30 MHz or more and 600 MHz or less.
Torr or less, a silicon-containing gas and hydrogen gas are used as source gases, and the ratio of the silicon-containing gas to the hydrogen gas is
It is characterized by being formed under the condition of not less than 0.5% and not more than 30%.

The electromagnetic wave having the above-mentioned frequency is converted into RF by plasma CVD (13.56 MHz is used industrially).
Since plasma can be generated at a lower pressure than the plasma CVD method, generation of polysilane in the gas phase can be eliminated,
A high-quality microcrystalline silicon-based semiconductor material can be formed.

Further, since plasma can be generated at a low pressure, the plasma can be expanded, which is very suitable for manufacturing a large-area photoelectric conversion element. In addition, since the deposition rate can be increased for these reasons, the throughput is increased, which is industrially advantageous. In addition, since the silicon-containing gas is diluted with a large amount of hydrogen gas, the supply of hydrogen-containing radicals to the film-forming surface is larger than in the case of forming an ordinary amorphous silicon-based semiconductor thin film. A microcrystalline silicon-based semiconductor thin film can be formed.

Further, since a negative self-bias is usually generated in the discharge electrode, it is possible to suppress the irradiation of the film formation surface with high-energy positive ion species.
A high-quality microcrystalline silicon-based semiconductor thin film can be formed.

Although the reason is unknown, a microcrystalline structure as in the present invention can be formed with good reproducibility by using such a method. Further, since the gas decomposition efficiency is higher than that of the RF plasma CVD method, the gas use efficiency is excellent, which is industrially advantageous.

(8) Method for Forming Second i-Layer Examples of the material for the second i-layer include amorphous silicon-based semiconductor materials, such as a-Si, a-SiC, and a-SiO. Especially
a-Si is excellent. Also, to make the i-layer more intrinsic,
Etc. may be added. H, C for compensating for dangling hands
It is desirable that the concentrations of l and F atoms are 0.1% or more and 10% or less. In order to form this layer, a plasma CVD method is usually used. Among them, the RF plasma CVD method is preferable. Deposition rate is 1A / sec or more, 20A / sec or less, formation temperature is 150 ° C or more, 35
It is desirable that the temperature is 0 ° C. or less and the pressure is 0.1 Torr or more and 5 Torr or less. Particularly, when forming a doped layer having a microcrystalline structure, a silicon-containing gas and a hydrogen gas are used as source gases,
The ratio of silicon-containing gas to hydrogen gas is 2% or more, 50
% Is desirable.

(9) At least one of the first doped layer 103, the second doped layer 105, the third doped layer 110, and the fourth doped layer 112 is made of a microcrystalline silicon semiconductor material. It is characterized by.

When a microcrystalline silicon-based semiconductor material is used for the doped layer, the carrier density of the layer can be increased, so that the open-circuit voltage of the photoelectric conversion element is improved. Furthermore, since the microcrystalline silicon-based semiconductor material has a smaller absorption coefficient in the visible light region than that of the amorphous silicon-based semiconductor material,
When used as a window layer on the light incident side, the short-circuit current increases.

Further, when a microcrystalline silicon semiconductor material is used for the first doped layer 103 and the second doped layer 105, there is no abrupt change at the interface with the first i-layer 104, so that the interface state becomes low. The amount is small and the fill factor of the photoconductive property is improved.

(10) As shown in FIG. 5, the first doped layer is formed as a layer (503) made of an amorphous silicon semiconductor material on the lower conductive layer side.
It is preferable to have a laminated structure of a) and a layer (503b) made of a microcrystalline silicon-based semiconductor material on the first i-layer side. The third doped layer may have a similar configuration. The fill factor of the photoelectric conversion element can be improved by forming the doped layer in a laminated structure as described above.

(11) The method of forming the transparent conductive layer of the present invention is made of a material selected from zinc oxide, tin oxide, indium oxide, ITO and zinc sulfide. However, zinc oxide or tin oxide is desirable in terms of high plasma resistance, easy control of the surface shape, and cost.

Using a DC magnetron sputtering method having a high deposition rate, the film is usually formed at a deposition rate of 10 (Å / sec) or more and 200 (Å / sec) or less. It is important to form at a temperature of 100 ° C. or more and 500 ° C. or less. Particularly, a temperature of 150 ° C. or more and 400 ° C. or less is preferable. At such a deposition rate and a forming temperature, a transparent conductive layer having a cross-sectional shape as in the present invention can be obtained, and the transmittance becomes 90% or more with light of 500 nm or more. Further, in order to form unevenness, after the layer is formed by the above method, the substrate surface may be appropriately etched using an acidic solution such as HNO3, HF, HCl, or H2SO4.

Hereinafter, other components will be described.

(Substrate) As the substrate 101, metal, resin, glass, ceramics, semiconductor bulk, and the like are used. The surface may have fine irregularities. In addition, by adopting a long shape, continuous film formation can be supported. Particularly, stainless steel, polyimide, and the like are preferable because they have flexibility.

(Reflection Layer) The reflection layer 102a has a role as an electrode and a role as a reflection layer that reflects light reaching the substrate and reuses it in the semiconductor layer. Al, Cu, Ag, Au
Are formed by a method such as vapor deposition, sputtering, plating, and printing.

The surface having irregularities has the effect of extending the optical path length of the reflected light in the semiconductor layer and increasing the short-circuit current.

When the substrate has conductivity, the reflective layer need not be formed.

(Upper Transparent Electrode) The upper transparent electrode 106 can also function as an antireflection film by appropriately setting its film thickness.

The transparent electrode 106 is formed of a material such as ITO, ZnO, InO3, etc. by using a method such as vapor deposition, CVD, spray, spin-on, and immersion. These compounds may contain a substance that changes electric conductivity.

(Current Collecting Electrode) The current collecting electrode 107 is provided to improve current collecting efficiency. As a forming method, there are a method of forming a metal of an electrode pattern by sputtering using a mask, a method of printing a conductive paste or a solder paste, and a method of fixing a metal wire with a conductive paste. FIG. 3 shows an example using a copper wire. A silver cladding layer is formed around a thin copper wire. This layer has a function of reducing the contact resistance with the copper wire. Further, a layer of carbon paste using an acrylic resin as a binder is formed around the silver clad layer. This layer has the function of maintaining the adhesion to the upper transparent electrode and the function of reducing the contact resistance with the silver clad layer. It also has a function of preventing silver in the silver cladding layer from diffusing into the photovoltaic layer.

Further, a bus bar or the like for taking out power is formed as shown in FIG. A plurality of collecting electrodes are arranged on the surface of the element without crossing, and one end of the collecting electrode is brought into electrical contact with the bus bar of 108. The busbar 108 is formed on the 107 and uses a metal material having good conductivity such as a Cu plate. Also, an insulating double-sided tape is disposed between the bus bar and the upper transparent electrode to make it adhere to the upper transparent electrode.

FIG. 4 shows an example of modularizing the photoelectric conversion element of the present invention. A plurality of photoelectric conversion elements are serialized as shown in FIG.
In the case where one photoelectric conversion element is shaded, all voltages generated from other photoelectric conversion elements are not applied to this photoelectric conversion element. Further, since the photoelectric conversion module of the present invention is sealed with the fluororesin and the support substrate after arranging the respective members as shown in FIG. 4, the invasion of water vapor can be suppressed.

[0057]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be specifically described below by taking a solar cell as an example of a photoelectric conversion element, but the present invention is not limited to this.

Example 1 The solar cell of FIG. 1A having one pin junction was manufactured. Specifically, the substrate stainless steel (SU
S430 10 × 10cm2 0.2mm thick) / reflective layer Ag / transparent conductive layer
ZnO / first doped layer a-Si: H: P / first i layer μc-Si: H
/ Second doped layer μc-Si: H: B / Upper transparent electrode ITO / Current collector electrode A solar cell composed of Cu wire / Ag / C (FIG. 3) was produced.

Here, the reflective layer and the transparent conductive layer are formed by the sputtering method using the apparatus shown in FIG. 8, the photovoltaic layer is formed using the apparatus shown in FIG. 9, and the first i-layer is a high frequency of 500 MHz. , A doped layer was formed by an RF plasma CVD method, and an upper transparent electrode layer was formed by a sputtering method.

The procedure for forming the lower conductive layer, the reflective layer, and the transparent conductive layer using the sputtering method will be described below. Fig. 8 shows an apparatus that can perform sputter etching and DC magnetron sputtering. 801 is a cylindrical deposition chamber, 802 is a substrate holder, 803 is a substrate, 804 is a heater, 805 is a matching box, 806 is an RF power supply, and 807 is a reflection. Metal target for forming a layer, 808 is a target for forming a transparent conductive layer, 810, 81
1 is DC power supply, 813 and 814 are shutters, 816 is exhaust pipe, 817
Is a gas introduction pipe, 818 is a rotation axis, and 819 is a plasma. In addition, although not shown, there are a gas supply device connected to the gas introduction pipe 817 and a vacuum pump connected to the exhaust pipe 816. An arrow 821 indicates the exhaust direction. First, the acid-washed and organic-washed substrate 803 is attached to a disk-shaped substrate holder, and a rotation shaft 818, which is the central axis of the disk-shaped substrate holder, is rotated. The inside of the deposition chamber was evacuated to 5 × 10E-6 Torr using an oil diffusion pump / rotary pump (not shown), Ar was introduced from a gas introduction pipe, and RF power was supplied from RF power source 806.
Electric power is introduced into the deposition chamber to generate Ar plasma. 80
Adjust the matching circuit in Step 5 to minimize the reflected power. At this time, the substrate is sputter-etched (reverse-sputtered) to have a more clean surface. Next, the heater is set to the temperature at which the reflective layer is formed. When the temperature reaches the predetermined temperature, the DC power supply of 810 is turned on, Ar plasma 819 is generated, the shutter 813 is opened, and the reflective layer has a predetermined film thickness. Once only formed, close the shutter and turn off DC power. next,
Set the heater so that it becomes the formation temperature of the transparent conductive layer,
When the temperature reaches a predetermined temperature, the DC power supply of 811 is turned on, Ar plasma is generated, the shutter 814 is opened, and when the transparent conductive layer is formed to a predetermined thickness, the shutter is closed and the DC power supply is turned off.

FIG. 9 shows an apparatus capable of carrying out a plasma CVD method. Reference numeral 901 denotes a reaction chamber; 902, a substrate on which a lower conductive layer is formed; 903, a heater; 904, a conductance valve;
8 is a high-frequency electrode, 909 is a high-frequency power supply (500 MHz) with a built-in matching circuit, 910 is plasma, 911 is a shutter, 914
Denotes an exhaust pipe, and 915 denotes a gas introduction pipe. 913 is the exhaust direction, 91
6 indicates the gas introduction direction. Although not shown in the figure, a vacuum pump such as an oil diffusion pump / rotary pump is connected to the exhaust pipe in the figure, and the gas introduction device is connected to the gas introduction pipe in the figure. The plasma CVD apparatus is configured as described above.

In order to actually form a layer using this plasma CVD apparatus, the following procedure is performed. First, the substrate 902 on which the lower conductive layer is formed is heated by the heater 90 inside the reaction chamber 901.
3 and evacuate with a vacuum pump such as an oil diffusion pump / rotary pump so that the pressure inside the reaction chamber is 1 × 10E-4 Torr or less. When the pressure becomes 1 × 10E-4 Torr or less, gases such as H2 and He are introduced into the reaction chamber from the gas introduction pipe 915, a heater is turned on, and the substrate 902 is set to a desired temperature. When the temperature of the substrate is stabilized, a source gas is introduced from a gas introduction tube, a high-frequency power source 909 is turned on, and high-frequency power is introduced from a high-frequency electrode 908 into the reaction chamber. The conductance valve 904 is adjusted so that a desired pressure is obtained when the plasma 910 is generated. At this time, it is preferable to adjust the matching circuit so that the reflected power is minimized. Next, the shutter 911 is opened, and when a layer having a desired film thickness is formed, the shutter is closed. The introduction of high-frequency power and the introduction of source gas are stopped, and preparations are made to form the next layer. To perform RF plasma CVD with this device, connect an RF power supply (13.56 MHz) instead of the high-frequency power supply 909, and
The plasma may be generated by applying electric power.

Table 1 shows specific conditions.

[0064]

[Table 1]

As shown in FIG. 13, the upper transparent electrode (ITO)
Apply an insulating double-sided tape to one side of the substrate
Cu wire shown in 3, Ag clad layer, one end of the collecting electrode made of carbon paste is fixed to the double-sided tape, and further, a bus bar is fixed to the double-sided tape from above the collecting electrode.
The whole was heated to fuse the carbon paste, and the current collecting electrode and the bus bar were fixed.

Several solar cells (actual 1) were manufactured. Transmission electron microscope (TEM)
As a result, it was found that the first i-layer had a structure in which the longitudinal direction of the columnar crystal grains shown in FIG. 1 was inclined with respect to the normal direction of the substrate. Also the lower conductive layer
When the surface roughness Ra of the (transparent conductive layer and reflective layer) was measured, it was found that the average per 50 μm length was 0.32 μm. The ratio of the angle G (the angle between the normal to the minute region of the lower conductive layer and the normal to the main surface of the substrate) of 15 degrees or more and 45 degrees or less was 91%. The ratio K of the total volume of the columnar crystal grains whose angle F is 20 degrees or less to the total volume of the i-layer is 90.
%was.

Comparative Example 1 As a lower conductive layer, only a substantially flat Ag was formed using a normal sputtering method without providing a transparent conductive layer, and a solar cell having the cross-sectional shape of FIG. 2 was manufactured. In addition, the first i-layer was formed using RF plasma CVD as shown in Table 1.
It formed under the conditions shown in b. Otherwise, a solar cell (the photoelectric conversion element in FIG. 2) similar to that in Example 1 was produced.

Table 2 shows the forming conditions.

[0069]

[Table 2]

Several pieces of the same solar cell (comparative 1) were produced. Transmission electron microscope (TEM)
As a result, it was found that columnar microcrystalline silicon was formed in the first i-layer as shown in FIG. Also, when the surface roughness Ra of the lower conductive layer was measured, it was found that the average per 50 μm length was 0.02 μm.

First, the solar cell of Example 1 (actual 1) and the comparative example 1
(Ratio 1) of the initial characteristics of the solar cell (photoconductive characteristics, leakage current,
(Low illuminance open voltage) was measured.

Solar simulator (AM1.5 100mW / cm2
When the photoelectric conversion efficiency, open-circuit voltage, and short-circuit photocurrent were measured using a surface temperature of 25 ° C.), the photoelectric conversion device of the present invention was 1.29 times, 1.04 times, and 1.23 times superior, respectively.

Next, when the open-circuit voltage was measured under a fluorescent lamp with an illuminance of about 500 lux (low illuminance), it was found that the solar cell of (actual 1) was 1.2 times better than that of (ratio 1). I understand. When a reverse bias was applied in a dark place and the leak current was measured, the leak current of the solar cell (actual 1) of the present invention was as excellent as about 1/8 of the comparative example 1 (ratio 1), which was excellent. I understand.

Next, a light irradiation test was performed on the solar cells of (actual 1) and (ratio 1). The above simulator (AM1.5 100mW / cm2
Exposure to a surface temperature of 50 ° C.) for 1,000 hours revealed no appearance defects after the test.

The photoelectric conversion efficiency after the light irradiation test, the open-circuit voltage,
When the short-circuit photocurrent, the open voltage at low illuminance, and the leak current were measured, differences were found in the decrease in the open voltage at low illuminance and the leak current before and after the test.

The open voltage ratio of low illuminance before and after the test (low illuminance open voltage after the test / low illuminance open voltage before the test) is
It was 0.95 for (actual 1) and 0.92 for (ratio 1). Furthermore, the ratio of the leak current before and after the test (the leak current after the test / the leak current before the test) was 1.2 in (actual 1) and 2.2 in (ratio 1).

As described above, the photoelectric conversion device of the present invention was found to be superior to the conventional photoelectric conversion device.

Further, the aforementioned “torsion test” was performed. No appearance defect was found in the two solar cells after the test.

The photoelectric conversion efficiency after the torsion test, the open voltage,
When short-circuit photocurrent, open voltage at low illuminance, and leakage current were measured, photoelectric conversion efficiency, open voltage, open voltage at low illuminance,
There was a difference in the decrease in the leakage current before and after the test.

The ratio of the photoelectric conversion efficiency before and after the test (photoelectric conversion efficiency after test / photoelectric conversion efficiency before test) was 0.98 in (actual 1) and 0.70 in (ratio 1). The ratio of the open-circuit voltage before and after the test (open-circuit voltage after the test / open-circuit voltage before the test) is
It was 0.99 in (actual 1) and 0.84 in (ratio 1).

The open-circuit voltage ratio at low illuminance before and after the test
(Low illumination open voltage after test / Low illumination open voltage before test)
Was 0.96 for (actual 1) and 0.87 for (ratio 1).

Further, the ratio of the leak current before and after the test
(Leak current after test / leak current before test) was 1.1 in (actual 1) and 3.1 in (ratio 1).

As described above, the photoelectric conversion device of the present invention was found to be superior to the conventional photoelectric conversion device.

Similarly, a hail test described in JIS C 8917 relating to crystalline solar cells was performed. The ice ball was dropped 10 times uniformly under the conditions of a diameter of 25 mm and a final speed of 23 m / sec. After the test, the same evaluation as described above was performed, and it was found that the solar cell of (actual 1) was similarly superior to the solar cell of (ratio 1).

(Example 2) As an example of another embodiment, FIG.
Was manufactured. Specifically, the substrate stainless steel
(SUS430 10 × 10cm2 0.2mm thick) / Reflective layer Al / Transparent conductive layer ZnO / First doped layer a-Si: H: P / First i layer μc-S
i: H / second doped layer μc-Si: H: B / third doped layer aS
i: H: P / 2nd i-layer a-Si: H / 4th doped layer μc-Si: H: B
Several solar cells (actual 2) composed of the material of / top transparent electrode ITO / collecting electrode Cu wire / Ag / C were fabricated.

Table 3 shows the forming conditions.

[0087]

[Table 3]

When the cross section of this solar cell was observed with a TEM, it was found that the first i-layer had a microcrystalline structure as shown in FIG. 1-b. Lower conductive layer (transparent conductive layer / reflective layer)
The surface roughness Ra was measured, and it was found that the average per 50 μm length was 0.29 μm. In addition, the angle G (the angle between the normal to the small area of the lower conductive layer and the normal to the main surface of the substrate) is 15
When the ratio of not less than 45 degrees and not more than 45 degrees was counted, it was 93%. The ratio K of the total volume of columnar crystal grains having an angle F of 20 degrees or less to the total volume of the first i-layer was 94%.

(Comparative Example 2) A comparative example having a configuration as shown in FIG.
In the solar cell 1 (Table 1b), between the second doped layer and the upper transparent electrode, the same third doped layer as in Example 2, the second i-layer,
Four doped layers were stacked to produce several photoelectric conversion elements (ratio 2). When the cross section of this solar cell was observed by TEM,
It was found that the first i-layer had a microcrystalline structure as shown in FIG. The same measurements and tests as in Example 1 were performed, and it was found that the solar cell of (Example 2) was superior to the solar cell of (Comparative 2).

Example 3 A long sheet was used as a substrate, and a reflective layer and a transparent conductive layer were sequentially formed by a roll-to-roll method with high productivity. In addition, a roll-to-roll method was used for forming the photovoltaic layer and the upper transparent electrode. The details will be described below.

The apparatus 1000 shown in FIG. 10 is a thin film forming apparatus capable of continuously forming several thin films in a vacuum on the surface of a long sheet-like substrate 1001 having flexibility (flexibility). . 1001 is a long substrate having flexibility such as stainless steel, 1008 is a delivery roll wound around the substrate in a roll shape, 1009 is a take-up roll for winding the substrate, and 1002 is a portion for fixing the delivery roll inside. A vacuum vessel that can be connected to a vacuum pump 1016 such as a rotary pump via a pipe 1018. Similarly, the take-up roll 1009 is fixed to the vacuum container 1007, and a vacuum pump is connected.

A gas gate 10 is provided between the vacuum vessels 1002 and 1007.
A substrate passage called 21 and vacuum vessels 1003 and 10 for forming a desired thin film by DC magnetron sputtering.
04, 1005, and 1006 are connected as shown in FIG. A gas introduction pipe 1010 is connected to the gas gate 1021 as shown in the figure so that a scavenging gas 1011 such as Ar flows into the gas gate 1021 so that mutual diffusion of gases does not occur between vacuum containers forming different types of thin films. Can be. Therefore, good bonding can be formed. The gas gate is connected between the vacuum vessels. However, when the same thin film is formed in a continuous vacuum vessel, it is not necessary to provide a gas gate between them.

[0093] Piping is applied to the vacuum vessels 1003, 1004, 1005, and 1006.
A diffusion pump 1017 is connected through 1019, and a vacuum pump such as a rotary pump is connected through piping. Further, inside the vacuum vessels 1003, 1004, 1005, and 1006, a heater 1014 for heating the substrate, a target 1023 for forming a desired thin film, an electrode 1013 with a built-in magnet, and a gas for introducing a gas 1022 for sputtering. Introductory tube 10
20 are provided. A DC power supply 1012 is connected to each electrode.

The method of using this device will be described below. First, a delivery roll 1008 wound around a flexible long substrate such as stainless steel is fixed in the vacuum container 1002, and the substrate tip is fixed inside the vacuum container 1007 through each gas gate, vacuum containers 1003, 1004, 1005, 1006. Take-up roll 1009
Wrap around.

[0095] Each vacuum pump is started up and evacuated until the internal pressure of each vacuum vessel becomes several mTorr. Gas inlet pipe 1010 to A
As for r gas, a desired gas is introduced from 1020, each heater is turned on, and the substrate is transported in the direction of arrow 1024. When the substrate temperature becomes constant, turn on each DC power supply and
Generate plasma 1015 in 1003, 1004, 1005, 1006,
Form a desired thin film.

When the substrate reaches the end of the substrate, the transport is stopped, each DC power supply and each heater power supply are turned off, and the substrate is cooled. When the temperature of the substrate becomes about room temperature, leak each vacuum container and take out the take-up roll.

Using a stainless steel substrate (SUS430) having a thickness of 0.15 mm, a reflective layer and a transparent conductive layer were continuously formed by the above method under the conditions shown in Table 3. This lower conductive layer (transparent conductive layer /
When the cross-sectional shape of the (reflective layer) was observed by SEM, it was found that the cross-sectional shape was as shown in FIG. The average surface roughness Ra per 50 μm length was 0.35 μm. Angle G is 15
When the ratio of not less than 45 degrees and not more than 45 degrees was counted, it was 90%.

Next, an apparatus for forming a photovoltaic layer on a transparent conductive layer by using a roll-to-roll method will be described in detail. Figure
The device 11 is a device for continuously forming six photovoltaic layers on a long substrate, and some vacuum vessels are omitted from the figure. 1101 is a long substrate having a lower conductive layer formed thereon, 1108 is a delivery roll wound around the substrate, 1109 is a take-up roll for winding the substrate, 1102 is a vacuum capable of fixing the delivery roll inside. In containers, plumbing 1118
And a vacuum pump 1116 such as a rotary pump. Similarly, the take-up roll 1109 is a vacuum container 1107
And a vacuum pump is connected. Vacuum container 11
Between 02 and 1107, a vacuum vessel 1103 for forming a desired thin film
-a, 1104, 1103-b (not shown), 1103-c (not shown), 1103-d,
1103-e are sequentially arranged, and a gas gate 11 is provided between each vacuum vessel.
21 is connected. As shown in the figure, a gas introduction pipe 1110 is connected to each gas gate 1121 to connect a scavenging gas 11 such as Ar, H2, and He.
11 can be introduced to prevent gas diffusion between the vacuum vessels forming different types of thin films. For this reason, the pin junction is very good.
When the same thin film is formed in a continuous vacuum vessel, it is not necessary to provide a gas gate therebetween.

Vacuum containers 1103-a, 1103-b, 1103-c, 1103-
d, 1103-e: RF plasma CVD method (power frequency 13.5
6 MHz). Piping 111
A rotary pump and a gas introduction pipe 1120 are connected through 8, and a heater 1114 and an RF electrode 1113 are fixed inside. An RF power supply 1112 is connected to the RF electrode. Also,
In the vacuum chamber 1104, a high-frequency plasma CVD method (power supply frequency: 150 MHz) can be performed. A diffusion pump 1117 is connected to the vacuum vessel through a pipe 1119, and a vacuum pump such as a rotary pump is connected through the pipe. Further, a gas introduction pipe is connected, and a heater 1127 and a high-frequency electrode 1126 are fixed inside. A high frequency power supply 1125 (frequency 150 MHz) is connected to the high frequency electrode.

The method of using this device will be described below. First, the delivery roll 1108 wound around the stainless steel substrate formed up to the lower conductive layer is fixed in the vacuum vessel 1102, and the substrate tip is passed through each gas gate and each vacuum vessel to form the vacuum vessel 1107.
Wrap around the take-up roll 1109 fixed inside. Activate each vacuum pump and evacuate until the internal pressure of each vacuum container becomes several mTorr. An H2 gas is introduced from the gas introduction pipe 1110, and a gas for forming a photovoltaic layer is introduced from the gas introduction pipe 1120, each heater is turned on, and the substrate is transported in the direction of arrow 1124. When the temperature of the substrate becomes constant, each RF power supply and high frequency power supply are turned on, matching is adjusted, and plasma 1115 is generated inside each vacuum vessel to form a desired thin film. When the substrate is reached at the end of the substrate, stop the transfer, turn off each DC power supply, each microwave power supply, DC power supply and each heater power supply, and cool the substrate. When the temperature of the substrate becomes about room temperature, leak each vacuum container and take out the take-up roll. Under the conditions shown in Table 3 using the method described above, the first doped layer a-Si: H:
P / 1st i layer μc-Si: H / 2nd doped layer μc-Si: H: B /
A third doped layer a-Si: H: P / a second i-layer a-Si: H / a fourth doped layer μc-Si: H: B was formed.

An upper transparent electrode (ITO) was continuously formed on the rolled solar cell taken out using the apparatus shown in FIG.
At that time, ITO was used as the target inside the vacuum vessel 1006, and plasma was not generated in the vacuum vessels 1003, 1004, and 1005, and the upper transparent electrode made of ITO was formed on the fourth doped layer only with the vacuum vessel 1006. Table 4 shows the conditions for forming each layer.

[0102]

[Table 4]

The rolled solar cell taken out is 30 × 30 cm
Then, as shown in FIG. 1-a, the same current collecting electrode and bus bar as in Example 1 were attached as shown in FIG. 1-a, and four solar cells were serialized as shown in FIG. 4, and each solar cell was connected in parallel. A bypass diode was connected. Next, the EV is placed on a 0.3 mm thick support substrate.
A, nylon resin, EVA, glass non-woven fabric, serialized solar cell, glass non-woven fabric, EVA, glass non-woven fabric, EVA, glass non-woven fabric, and fluororesin were laminated and heated and vacuum-sealed (laminated).

When the same measurement and test as in Example 1 were performed on the solar cell module (actual 3) having a size of 35 × 130 cm 2 manufactured as described above, any of the initial characteristics, the torsion test, and the hail test was performed. In addition, the characteristics were even better than the solar cell of (actual 1). The solar cell module
(Result 3) was left outdoors for 3 months, and an outdoor exposure test was performed.
Almost no change in appearance was observed, and the decrease in photoelectric conversion efficiency was about 5%.

As described above, it was found that the photoelectric conversion module of the present invention had excellent characteristics.

Example 4 A solar cell (Example 4) was produced in the same manner as in Example 1 except that the thickness of the first i-layer was changed to 2 μm. Observation of the cross section with a TEM revealed that the structure of the first i-layer was as shown in FIG. The same measurement and test as in Example 1 were performed, and it was found that the solar cell of (Example 4) was as excellent as the solar cell of (Example 1).

(Example 5) As an example of another embodiment, a solar cell in which the lower conductive layer was formed of a single layer was manufactured. The lower conductive layer was formed of a 0.5 μm-thick silver layer and was formed by a sputtering method. The surface shape of the layer was made uneven (textured) by setting the formation temperature to 350 ° C. By observing the cross section, the ratio of the angle G (the angle formed by the normal to the minute region of the lower conductive layer and the normal to the main surface of the substrate) of 15 degrees or more and 45 degrees or less was counted and found to be 91%. The same photovoltaic layer, upper transparent electrode, and current collecting electrode as in Example 1 were formed on the lower conductive layer, and several solar cells (Example 5) of FIG. 1 were manufactured. The same measurement and test as in Example 1 were performed, and it was found that the solar cell of (Example 5) was as excellent as the solar cell of (Example 1).

Example 6 As another example of the embodiment, the photovoltaic layer shown in FIG. 5 in which the first doped layer has a laminated structure of a-Si: H: P / μc-Si: H: P The solar cell shown in FIG. In Example 2, the first doped layer was a-Si: H: P / μc-S
A solar cell (Example 6) was produced in the same manner as in Example 2 except that i: H: P was used. The same measurements and tests as in Example 1 were performed, and it was found that the solar cell of (Example 4) was as excellent as the solar cell of (Example 2).

[0109]

According to the photoelectric conversion device of the present invention, characteristics such as photoelectric conversion efficiency, open-circuit voltage, short-circuit photocurrent, low-illuminance open-circuit voltage, and leak current of the photoelectric conversion device can be improved. It can also improve the outdoor exposure test, mechanical strength, and durability in long-time light irradiation. Further, the cost of the photoelectric conversion element can be significantly reduced.

[Brief description of the drawings]

FIG. 1 shows a photoelectric conversion element of the present invention characterized by a lower conductive layer and an i-layer.

FIG. 2 shows a conventional photoelectric conversion element

FIG. 3 shows an example of a collecting electrode

FIG. 4 shows an example of modularization of the photoelectric conversion element of the present invention.

FIG. 5 is a photoelectric conversion device of the present invention characterized by a first doped layer.

FIG. 6 shows the relationship between the surface roughness of the lower conductive layer and the photoelectric conversion efficiency.

FIG. 7 shows the relationship between the ratio at which the angle G defined in FIG. 1a is 15 degrees or more and 45 degrees or less and the photoelectric conversion efficiency

FIG. 8 shows an apparatus for forming a lower conductive layer and an upper transparent electrode of the photoelectric conversion element of the present invention.

FIG. 9 shows an apparatus for forming a photovoltaic layer of a photoelectric conversion element according to the present invention.

FIG. 10 shows an apparatus for continuously forming a lower conductive layer and an upper transparent electrode of the photoelectric conversion element of the present invention.

FIG. 11 is an apparatus for continuously forming a photovoltaic layer of a photoelectric conversion element according to the present invention.

12A is a diagram illustrating a relationship between the ratio of the angle F defined in FIG. 12B of 0 degree or more and 20 degrees or less and the photoelectric conversion efficiency, and b is a diagram defining the angle F.

FIG. 13 shows an example in which a bus bar is provided in the photoelectric conversion element of the present invention.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 101 Substrate 102 Lower conductive layer 102a Reflective layer 102b Transparent conductive layer 103 1st dope layer 104 1st i-layer 105 2nd dope layer 106 Upper transparent electrode 107 Collector electrode 108 Bus bar 109 Double-sided tape 110 3rd dope layer 111 Second i-layer 112 Fourth doped layer 120 Angle G 201 Substrate 202 Lower conductive layer 203 First doped layer 204 First i-layer 205 Second doped layer 206 Upper transparent electrode 207 Collector electrode 301 Copper wire 302 Silver clad layer 303 Carbon paste layer 306 Surface of upper transparent electrode 307 Current collecting electrode 401 Support substrate 402, 404, 409, 411 EVA 403 Nylon resin 405, 408, 410, 412 Glass nonwoven fabric 406 Bypass diode 407 Photoelectric conversion element 413 Fluorine Resin 14 Bus bar 503a First doped layer a 503b First doped layer b 504 First i layer 505 Second doped layer 506 Third doped layer 507 Second i layer 508 Fourth doped layer 801 Deposition chamber 802 Substrate holder 803 Substrate 804 Heater 805 Matching box 806 RF power source 807 Target for reflective layer 808 Target for transparent conductive layer 810, 811 DC power source 813,814 Shutter 816 Exhaust pipe 817 Gas inlet pipe 818 Rotation axis 819 Plasma 901 Reaction chamber 902 Substrate on which lower conductive layer is formed 903 Heater 904 Conductance valve 908 High frequency electrode 909 High frequency power supply 910 Plasma 911 Shutter 913 Exhaust direction 914 Exhaust pipe 915 Gas introduction pipe 916 Gas introduction direction 1001 Long substrate 1002, 1003, 1004, 1005, 1006
1007 Vacuum container 1008 Delivery roll 1009 Take-up roll 1010 Gas introduction pipe 1011 Scavenging gas 1012 DC power supply 1013 Electrode 1014 Heater 1015 Plasma 1016 Vacuum pump 1017 Diffusion pump 1018, 1019 Exhaust pipe 1020 Gas introduction pipe 1021 Gas gate 1022 Gas 102 Transport direction 1101 Long substrate 1102 with lower conductive layer formed 1102, 1103-a, 1103-b, 1103-
c, 1103-d, 1103-e, 1103-f, 11
04,1107 Vacuum container 1108 Delivery roll 1109 Take-up roll 1110 Scavenging gas introduction pipe 1111 Scavenging gas 1112 RF power supply 1113 RF electrode 1114 Heater 1115 Plasma 1116 Vacuum pump 1117 Diffusion pump 1118,1119 Exhaust pipe 1120 Gas introduction pipe 1122 Gas gate 1122 1124 Substrate transfer direction 1125 High frequency power supply 1126 High frequency electrode 1127 Heater

Claims (12)

[Claims] A substrate, a lower conductive layer, a first doped layer, i
Layer, a second doped layer, and in the photoelectric conversion element having an upper conductive layer, the lower conductive layer surface has an uneven shape,
A photoelectric conversion element, wherein the i-layer contains columnar crystal grains, and a longitudinal direction of the columnar crystal grains is arranged to be inclined with respect to a normal direction of the substrate. 2. A straight line A passing through the columnar crystal grains and parallel to a longitudinal direction thereof, an interface 1 between a first doped layer and the i-layer, and a second
Among the straight lines connecting the shortest distance between the doped layer and the interface 2 of the i-layer, the ratio of the angle formed by the straight line B passing through the columnar crystal grains A to 20 degrees or less is 70% with respect to the entire volume of the i-layer. A photoelectric conversion element characterized by the above. 3. The method according to claim 1, wherein the lower conductive layer located between the substrate and the semiconductor layer has a surface roughness Ra of 0.1 μm at a length of about several tens μm.
2. The photoelectric conversion device according to claim 1, wherein the size is not less than m and not more than 1 μm. 4. A region in which the normal direction of the surface of the lower conductive layer in the minute region of the lower conductive layer is 15 ° or more and 45 ° or less with respect to the normal of the main surface of the substrate is the entire surface region. 2. The photoelectric conversion device according to claim 1, wherein the content is 80% or more. 5. The method according to claim 1, wherein the length of the columnar crystal grains in the longitudinal direction is 100.
2. The photoelectric conversion element according to claim 1, wherein the thickness is not less than Angstroms and not more than 0.3 μm. 6. The photoelectric conversion element according to claim 1, wherein the thickness of the i-layer is 0.3 μm or more and 3 μm or less. 7. The method according to claim 1, wherein a minute region made of an amorphous silicon-based semiconductor material exists in the i-layer at a volume ratio of 50% or less with respect to the entire region of the i-layer. The photoelectric conversion device according to any one of the preceding claims. 8. The method according to claim 1, wherein the i-layer is formed by a plasma CVD method using an electromagnetic wave having a frequency of 30 MHz or more and 600 MHz or less, and a pressure of 1 mTorr or more and 1 Torr or less. 2. The photoelectric conversion element according to claim 1, wherein the photoelectric conversion element is formed under a condition that a ratio of the silicon-containing gas to the silicon is 0.5% or more and 30% or less. 9. A semiconductor device comprising a third doped layer, a second i-layer, and a fourth doped layer between the second doped layer and the upper conductive layer, wherein the second i-layer is non-conductive. 2. The photoelectric conversion element according to claim 1, comprising a crystalline silicon-based semiconductor, and having a thickness of 0.1 μm or more and 0.4 μm or less. 10. The semiconductor device according to claim 1, wherein at least one doped layer includes a microcrystalline silicon-based semiconductor material.
9. The photoelectric conversion element according to 9. 11. The microcrystalline silicon-based semiconductor device according to claim 1, wherein the first and / or third doped layer has a laminated structure of a layer made of a microcrystalline silicon-based semiconductor material and a layer made of an amorphous silicon-based semiconductor material. A photoelectric conversion element, wherein a layer made of a semiconductor material is in contact with the i-layer. 12. The photoelectric conversion device according to claim 1, wherein the substrate is a long strip-shaped substrate.